While a number of recent studies have focused on the role of stromal cells in advanced carcinomas, the stromal changes contributing to carcinogenesis have received considerably less attention. Gastric and hepatocellular cancer are two of the leading causes of cancer mortality worldwide, and are strongly linked to chronic infection (H. pylori versus HBV/HCV, respectively) and the consequent state of sustained inflammation. This application is submitted by the Tumor Microenvironment Program at the Columbia University Cancer Center, bringing together leading investigators in the areas of gastric carcinogenesis and liver fibrosis with experts in other areas to examine the role of stromal cells in liver and gastric carcinogenesis. This highly collaborative and multidisciplinary group encompasses experts in stromal cells, chronic inflammation/NF-kB, angiogenesis, epigenetics, optical imaging, and bioinformatics. It also includes a clinical arm that will conduct parallel studies on stromal cells from primary gastric and liver tumors, and a translational component (effects of COX-2 inhibition on tumor vascular stromal cells). Our proposal is based heavily on novel observations by this team of researchers regarding the bone marrow origin of cancer and stromal cells, the mechanism of activation of myofibroblasts (e.g. stellate cells), and on the link between chronic inflammation, angiogenesis, and the development of cancer. The program includes 4 projects and 2 cores. Project 1. Stromal myofibroblasts in gastric carcinogenesis (PI - Timothy C. Wang). Project 2. Stromal myofibroblasts in hepatic carcinogenesis (PI - David A. Brenner) Project 3. Interactions between digestive cancers and the vascular stroma (PI - Jan Kitajewski) Project 4. Epigenetics and genetics of stromal cells in liver and gastric cancer (PI - Benjamin Tycko) The Cores, which include an Imaging Core led by Andreas Hielscher, and an Administrative/Bioinformatics Core led by Timothy C. Wang and Andrea Califano, will be utilized by most of the projects and will provide additional focus and unity to the Project as a whole. In addition, most of the projects will utilize unique transgenic models of cancer (e.g. INS-GAS), reporter gene mice (ASMA-RFP/COLL-EGFP/Luc) targeting myofibroblasts, and a new method of methylation-sensitive SNP chip analysis (MSNP) to profile the myofibroblasts. Taken together, the data from these murine and human studies should lead to new insights regarding tumor-stroma interactions and their origins in early cancer.